Agentic Risk Analyst

What could go wrong with this AI use case, and what should we do about it?

Chirag Dani · ARA v3.2.1
LLM: huggingface · meta-llama/Llama-3.3-70B-Instruct · Secure-RAG: connected

Automated decision-making transparency pack

From 10 December 2026, an organisation covered by the Privacy Act must describe in its privacy policy the automated decisions that meet all three conditions of APP 1.7. List each system, answer three questions, and ARA applies the test, explains the result, drafts the APP 1.8 policy text and builds the evidence file.

0
Systems
0
In scope
0
Need legal review
0
Not in scope

Inventory

No systems yet. Add one, import a CSV, or create one from an ARA assessment.

Draft privacy policy text (APP 1.8)

Automated decisions

[Organisation] uses computer programs to make, or to help make, some decisions that could significantly affect you.

Kinds of personal information used in these programs: [list the kinds].

[Add how individuals can ask about or seek review of a decision, and contact details.]

APP 1.8 (commences 10 December 2026). Draft only: a qualified person must review before publication.

Import from CSV

Header row with any of: name, client, business_unit, description, decision, affected_individuals, role, effect, personal_information, pi_kinds, human_review, vendor, vendor_product, owner, review_date, evidence, notes, evidence_workspace. Role: makes, key_factor, minor, none or unsure. Effect: benefit, contract, service, other_significant, trivial or unsure. Personal information: yes, no or unsure.

Template

Decision support only, not legal or regulatory advice. AI-generated ratings and citations must be reviewed by a qualified person before they are relied on.